Anthropic's self-assessment: cheap misuse monitoring using the model itself
Source
safeguards.anthropic.com
Author
Anthropic
Date
Why it matters
Self-assessment reuses the cached conversation to flag misuse for about 10 extra tokenThe chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.Full definition →, adding under 2% compute. Teams building AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → monitoring get a cheap monitor design, plus evidence that it works offline but is weak as a blocker.
Terms in this piece · Glossary
token — The chunk of text a model reads and writes in — roughly three-quarters of a word — and the unit AI usage is billed in.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
